We propose to develop a model-driven, evidence-based pain item bank for use in pain assessment that is specific to geriatric oncology (gero-oncology) patients. This will build on our Novel Pain Assessment and Intervention Network (NoPAIN) project (Chang, PI, 1R21CA113191-01), and continue its Clinical Infometrics approach. This proposed study is designed in response to PA-03-152(Biobehavioral Pain Research) to "refine existing techniques for measuring pain and develop new techniques that are disease- and outcome- specific for different populations." We have chosen the condition of pain in this population because: pain is a complex multidimensional experience; a critical mass of pain instruments now exist and there is an urgent need to refine and implement them; and pain management in cancer among older adults is a pressing issue given societal demographics. We also aim to construct a computerized adaptive testing platform for this item bank that: 1) can reliably measure the multidimensional pain experience of heterogeneous cancer patients; and 2) is sensitive to change so that it can assess the effectiveness of treatments over time. Our four distinct but related specific aims are:1) To identify and refine the domains of pain assessment for a gero-oncology population, using existing theoretical frameworks in biopsychosocial medicine and palliative care as a guide; 2) To compile the items for a multidimensional pain item bank, drawing from existing pain questionnaires and supplementing with newly written items; 3) To develop empirically a pain item bank applicable to a gero-oncology population; and 4) To pilot test a computerized adaptive testing (CAT) platform to administer individualized pain assessments in clinical settings. At the end of this R21 project, we expect that (1) the conceptual model of the gero-oncology pain measurement can be improved; (2) the initial sets of pain items can be constructed; (3) items displaying differential item functioning (DIP) can be identified and revised; and (4) the CAT prototype will demonstrate how the system should be implemented and tested in clinical settings.